OpenAI Is Cashing In With ChatGPT Plus

02-02-2023

⚡ Today’s Highlights

  • 📰 News: OpenAI announces ChatGPT Plus and Microsoft is warning employees not to share sensitive info with ChatGPT

  • 💰 Funding: Lulalend, Phantom AI, Aluna, and Dxcover

  • 🦾 Tools: No more mysterious links and Grammarly for programmers

  • 📅 Events: The World AI Cannes Festival, RE•WORK AI Summit West, Data Science Salon Austin

📰 Today's Top Stories

(2 min read) (Source: OpenAI)

TLDR: ChatGPT Plus is a new monthly subscription plan for $20 that offers subscribers faster response times, priority access to new features and improvements, and general access to ChatGPT. It’s currently only available in the US, but plans to expand to other regions are in place. Free access to ChatGPT will continue to be offered.

ChatGPT was launched as a research preview and has received millions of user feedback, with users finding value in the system for various professional use cases. Future plans for ChatGPT include refining and expanding the offering based on feedback, launching the API waitlist, and exploring options for lower-cost plans, business plans, and data packs. Check out the announcement and discussion on Twitter.

The Big Picture: They had to monetize it at some point. The costs of processing that much data are probably eye-watering (even with the comfy cash infusion from Microsoft). With more companies incorporating ChatGPT and services like it into their operations, it’s just too hard to tell how fast and how high OpenAI will fly and you can bet they’re gonna need all the cash they can get.

(6 min read) (Source: Business Insider)

TLDR: Keep your enemies close, and your business partners closer? Microsoft, who recently signed a multi-billion dollar partnership with OpenAI, is warning employees to avoid sharing sensitive information with ChatGPT. Amazon issued a similar warning to its employees because accidentally sharing such data with ChatGPT could lead to privacy breaches in the future.

For obvious reasons, it would be quite difficult to identify and remove personal information from the growing ocean of data used by the chatbot once it’s lost there. OpenAI's terms of service gives them the right to use all inputs and outputs generated by users and ChatGPT, meaning that in the wrong hands, this kind of information could be used for corporate espionage.

The Big Picture: Companies (through some cleverly worded prompts) may be able to access private information and utilize it for nefarious purposes. The big question is, who's responsible for protecting confidential information and how can companies prevent it from being used for corporate espionage?

Semafor has also learned that Microsoft’s Bing is preparing to integrate a more advanced and feature-rich version of ChatGPT, called GPT-4, into its search engine in the near future. More data = more risks. Everyone using these services will have to be very careful with their personal information because you never know who might have access to it.

(10 min read) (Source: WIRED)

TLDR: Newtral, a Madrid-based media company, has developed "ClaimHunter," an AI language model that helps automate the work of fact-checkers. The system uses Microsoft's BERT model and was trained on 10,000 statements to recognize facts. It reduces the time to identify statements to be checked by 70-80% according to a journalist at Newtral.

The company is working with the London School of Economics and ABC Australia to develop a tool to identify repeated false statements. Despite the benefits, AI has limitations and can amplify biases and spread false information, making the job of fact-checkers even harder in a world where misinformation is becoming increasingly prevalent and resources are limited.

The Big Picture: The growing threat of dis- and mis- information has been greatly increased by generative AI tools that can craft convincing campaigns of false information in an instant. The demand for fact-checkers will be higher than ever and widespread job cuts across the market will likely increase the responsibility of independent organizations to detect and correct false information.

(8 min read) (Source: CNBC)

TLDR: Google is testing AI-powered chat products including Apprentice Bard, a chatbot powered by Google's LaMDA language model, and a question-and-answer search engine. The chatbot utilizes Google's LaMDA language model and has been pitted against OpenAI's ChatGPT. Internal documents show the performance of the two chatbots was compared, with Apprentice Bard demonstrating the ability to provide information about recent events, a feature that ChatGPT currently lacks.

The document compared their responses to a prompt asking if ChatGPT and AlphaCode would replace programmers, to which both chatbots answered no, stating that programming requires human expertise and creativity. However, the same internal documents showed that ChatGPT can demonstrate enough coding expertise to be hired as an entry-level coder at Google. It remains unclear if and when these experiments will be launched to the public, as Google executives have expressed concerns about the risk of disseminating misinformation.

The Big Picture: “Apprentice Bard” is the first step in Google’s response to ChatGPT and their move towards a more intelligent search engine. However, Google executives are hesitant to release an incomplete or potentially harmful product, stating that their company faces much higher reputational risk from the release of products that disseminate misinformation. Will they wait and see how ChatGPT handles the problem and what will this mean for delays on a project like this when Bing and GPT-4 are just around the corner?

💰 Funding Alerts

  1. Lulalend, a digital lender from South Africa is aiming to increase access to credit for small businesses in Africa. Fintech companies have been exploring the use of AI to reduce friction and bias in the loaning process to give better access to those that might be considered “less favourable” lendees by c. They recently raised $35 million in a Series B funding round, led by international impact investor Lightrock, and claim to use a proprietary blend of a credit scoring algorithm and a diverse set of alternative data sources to provide short-term loans in hours instead of days or months.

  2. Phantom AI, a startup located in Silicon Valley that specializes in autonomous driving technology, has announced the completion of a $36.5 million funding round in its Series C funding. The company's platform includes computer vision, sensor fusion and control capabilities.

  3. Aluna, a digital spirometer that measures the capacity and function of your lungs, has developed an AI-based respiratory management platform that allows patients to track and share their symptoms, medication, activity and environmental factors with their medical team. They recently secured $15.3 million in its Series B funding round to support patients suffering from asthma, and cystic fibrosis.

  4. Dxcover, a liquid biopsy platform based in Glasgow, uses infrared spectroscopy to analyze patient blood samples and AI algorithms to detect the presence of diseases. They raised £9.7 million in Series A investment and grant funding for the development of a blood test that can detect cancer in its early stages.

🦾 Trending Tools

🌎 Popular Content

1. Oh boy, here we go… Check out the rest of the thread and its discussion here. (Reddit) 

2. ChatGPT might have been the first to make it big but artificial intelligence is transforming the world and people can feel it. Check out the discussion here. (Reddit)

3. Speaking of the magic word machine, folks over on Reddit are discussing some of the ways that they’ve used ChatGPT to make money. (Reddit)

4. As someone who is in that 15-35 range and (obviously) pretty blown away by what AI has already done, I feel called out. Do you agree with Doug? (Twitter)

5. Is this wholesome or is it telling us what it thinks we want to hear… (Reddit)

6. Incredible, when are we giving ChatGPT its Broadway show? (YouTube)

👀 More Reading

🧠 Resources

Here are some of the resources shared by DeepMind to help people at different stages of their learning journey to find out more about AI and better employ it in their professional and personal lives. It includes a number of lectures covering different topics and Coursera specializations for undergrads.

 📅 Upcoming Events

  1. The World AI Cannes Festival (February 9-11, 2023. Cannes, France + Virtual), where decision-makers and AI innovators meet, where the most promising innovations and technologies get into the spotlight, where those who are currently building the world’s most game-changing AI strategies and use-cases will be on stages.

  2. RE•WORK AI Summit West, Deep Learning Summit (February 15-16, 2023. San Francisco, CA), a chance to hear the latest technology advancements, practical examples of how to apply AI to solve challenges in cross-industry settings, business and society, and delve deeper into the work of leading AI experts in a series of presentations, panel discussions, interviews and fireside chats.

  3. Data Science Salon Austin (February 21-22, 2023. Austin, TX + Virtual) is a two-day 500-person conference focused on AI and machine learning applications in the enterprise. The intimate event curates data science sessions to bring industry leaders and specialists face-to-face to educate each other on innovative new solutions in artificial intelligence, machine learning, predictive analytics and acceptance around best practices.

 💼 Jobs

Powered by AI. Curated and edited by Humans.

How'd you like today's email?

Login or Subscribe to participate in polls.

We'd love to hear from you! Leave us comments or feedback by replying to this email!